jyoung105/auraflow-v3 | Run with an API on Replicate

jyoung105 / auraflow-v3

AuraFlow: Fully open-sourced flow-based text-to-image generation model

Cold

Public
122 runs
L40S
GitHub
Weights

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

Run this model in Node.js with one line of code:

npx create-replicate --model=jyoung105/auraflow-v3

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";
import fs from "node:fs";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run jyoung105/auraflow-v3 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "jyoung105/auraflow-v3:4af944a25c343cf9c8a36f024ac120e6f66aa758b75f71061ddab9521049fbd3",
  {
    input: {
      steps: 50,
      width: 1024,
      height: 1024,
      prompt: "A man with hoodie on, illustration",
      num_images: 1,
      guidance_scale: 3.5
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run jyoung105/auraflow-v3 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "jyoung105/auraflow-v3:4af944a25c343cf9c8a36f024ac120e6f66aa758b75f71061ddab9521049fbd3",
    input={
        "steps": 50,
        "width": 1024,
        "height": 1024,
        "prompt": "A man with hoodie on, illustration",
        "num_images": 1,
        "guidance_scale": 3.5
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run jyoung105/auraflow-v3 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "jyoung105/auraflow-v3:4af944a25c343cf9c8a36f024ac120e6f66aa758b75f71061ddab9521049fbd3",
    "input": {
      "steps": 50,
      "width": 1024,
      "height": 1024,
      "prompt": "A man with hoodie on, illustration",
      "num_images": 1,
      "guidance_scale": 3.5
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/jyoung105/auraflow-v3@sha256:4af944a25c343cf9c8a36f024ac120e6f66aa758b75f71061ddab9521049fbd3 \
  -i 'steps=50' \
  -i 'width=1024' \
  -i 'height=1024' \
  -i 'prompt="A man with hoodie on, illustration"' \
  -i 'num_images=1' \
  -i 'guidance_scale=3.5'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/jyoung105/auraflow-v3@sha256:4af944a25c343cf9c8a36f024ac120e6f66aa758b75f71061ddab9521049fbd3
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "steps": 50,
      "width": 1024,
      "height": 1024,
      "prompt": "A man with hoodie on, illustration",
      "num_images": 1,
      "guidance_scale": 3.5
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

{
  "completed_at": "2024-11-07T03:36:51.533753Z",
  "created_at": "2024-11-07T03:29:41.896000Z",
  "data_removed": false,
  "error": null,
  "id": "7smhry5m11rgp0ck0jnv4stch8",
  "input": {
    "steps": 50,
    "width": 1024,
    "height": 1024,
    "prompt": "A man with hoodie on, illustration",
    "num_images": 1,
    "guidance_scale": 3.5
  },
  "logs": "DEVICE: cuda\nDTYPE: torch.float16\nUsing seed: 12780\nFinish setup in 0.0001506805419921875 secs.\n[Debug] Prompt: A man with hoodie on, illustration, best quality, high detail, sharp focus\n  0%|          | 0/50 [00:00<?, ?it/s]\n  2%|▏         | 1/50 [00:01<01:08,  1.39s/it]\n  4%|▍         | 2/50 [00:01<00:37,  1.28it/s]\n  6%|▌         | 3/50 [00:02<00:42,  1.11it/s]\n  8%|▊         | 4/50 [00:03<00:44,  1.04it/s]\n 10%|█         | 5/50 [00:04<00:44,  1.01it/s]\n 12%|█▏        | 6/50 [00:05<00:44,  1.01s/it]\n 14%|█▍        | 7/50 [00:06<00:43,  1.02s/it]\n 16%|█▌        | 8/50 [00:08<00:43,  1.03s/it]\n 18%|█▊        | 9/50 [00:09<00:42,  1.04s/it]\n 20%|██        | 10/50 [00:10<00:41,  1.04s/it]\n 22%|██▏       | 11/50 [00:11<00:40,  1.05s/it]\n 24%|██▍       | 12/50 [00:12<00:39,  1.05s/it]\n 26%|██▌       | 13/50 [00:13<00:38,  1.05s/it]\n 28%|██▊       | 14/50 [00:14<00:37,  1.05s/it]\n 30%|███       | 15/50 [00:15<00:36,  1.05s/it]\n 32%|███▏      | 16/50 [00:16<00:35,  1.06s/it]\n 34%|███▍      | 17/50 [00:17<00:34,  1.06s/it]\n 36%|███▌      | 18/50 [00:18<00:33,  1.06s/it]\n 38%|███▊      | 19/50 [00:19<00:32,  1.06s/it]\n 40%|████      | 20/50 [00:20<00:31,  1.06s/it]\n 42%|████▏     | 21/50 [00:21<00:30,  1.06s/it]\n 44%|████▍     | 22/50 [00:22<00:29,  1.06s/it]\n 46%|████▌     | 23/50 [00:23<00:28,  1.06s/it]\n 48%|████▊     | 24/50 [00:24<00:27,  1.06s/it]\n 50%|█████     | 25/50 [00:25<00:26,  1.05s/it]\n 52%|█████▏    | 26/50 [00:27<00:25,  1.05s/it]\n 54%|█████▍    | 27/50 [00:28<00:24,  1.05s/it]\n 56%|█████▌    | 28/50 [00:29<00:23,  1.05s/it]\n 58%|█████▊    | 29/50 [00:30<00:22,  1.05s/it]\n 60%|██████    | 30/50 [00:31<00:21,  1.05s/it]\n 62%|██████▏   | 31/50 [00:32<00:20,  1.05s/it]\n 64%|██████▍   | 32/50 [00:33<00:18,  1.05s/it]\n 66%|██████▌   | 33/50 [00:34<00:17,  1.05s/it]\n 68%|██████▊   | 34/50 [00:35<00:16,  1.05s/it]\n 70%|███████   | 35/50 [00:36<00:15,  1.05s/it]\n 72%|███████▏  | 36/50 [00:37<00:14,  1.05s/it]\n 74%|███████▍  | 37/50 [00:38<00:13,  1.05s/it]\n 76%|███████▌  | 38/50 [00:39<00:12,  1.06s/it]\n 78%|███████▊  | 39/50 [00:40<00:11,  1.06s/it]\n 80%|████████  | 40/50 [00:41<00:10,  1.06s/it]\n 82%|████████▏ | 41/50 [00:42<00:09,  1.06s/it]\n 84%|████████▍ | 42/50 [00:43<00:08,  1.06s/it]\n 86%|████████▌ | 43/50 [00:44<00:07,  1.06s/it]\n 88%|████████▊ | 44/50 [00:46<00:06,  1.06s/it]\n 90%|█████████ | 45/50 [00:47<00:05,  1.06s/it]\n 92%|█████████▏| 46/50 [00:48<00:04,  1.06s/it]\n 94%|█████████▍| 47/50 [00:49<00:03,  1.06s/it]\n 96%|█████████▌| 48/50 [00:50<00:02,  1.06s/it]\n 98%|█████████▊| 49/50 [00:51<00:01,  1.06s/it]\n100%|██████████| 50/50 [00:52<00:00,  1.06s/it]\n100%|██████████| 50/50 [00:52<00:00,  1.05s/it]\nFinish generation in 54.51199126243591 secs.",
  "metrics": {
    "predict_time": 56.405148416,
    "total_time": 429.637753
  },
  "output": [
    "https://replicate.delivery/pbxt/fpelzzm3O7qDRkVeVJsC84YfjvK7RZ0Bsv7t0ToDBZOL955OB/out_0.png"
  ],
  "started_at": "2024-11-07T03:35:55.128604Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/7smhry5m11rgp0ck0jnv4stch8",
    "cancel": "https://api.replicate.com/v1/predictions/7smhry5m11rgp0ck0jnv4stch8/cancel"
  },
  "version": "91da0bd735b6eb3a321656f78b1e96db694b49391c9f1140ad1134badc8a84ef"
}

Generated in

56.4 seconds

Tweak it ShareReport View full prediction

DEVICE: cuda
DTYPE: torch.float16
Using seed: 12780
Finish setup in 0.0001506805419921875 secs.
[Debug] Prompt: A man with hoodie on, illustration, best quality, high detail, sharp focus
  0%|          | 0/50 [00:00<?, ?it/s]
  2%|▏         | 1/50 [00:01<01:08,  1.39s/it]
  4%|▍         | 2/50 [00:01<00:37,  1.28it/s]
  6%|▌         | 3/50 [00:02<00:42,  1.11it/s]
  8%|▊         | 4/50 [00:03<00:44,  1.04it/s]
 10%|█         | 5/50 [00:04<00:44,  1.01it/s]
 12%|█▏        | 6/50 [00:05<00:44,  1.01s/it]
 14%|█▍        | 7/50 [00:06<00:43,  1.02s/it]
 16%|█▌        | 8/50 [00:08<00:43,  1.03s/it]
 18%|█▊        | 9/50 [00:09<00:42,  1.04s/it]
 20%|██        | 10/50 [00:10<00:41,  1.04s/it]
 22%|██▏       | 11/50 [00:11<00:40,  1.05s/it]
 24%|██▍       | 12/50 [00:12<00:39,  1.05s/it]
 26%|██▌       | 13/50 [00:13<00:38,  1.05s/it]
 28%|██▊       | 14/50 [00:14<00:37,  1.05s/it]
 30%|███       | 15/50 [00:15<00:36,  1.05s/it]
 32%|███▏      | 16/50 [00:16<00:35,  1.06s/it]
 34%|███▍      | 17/50 [00:17<00:34,  1.06s/it]
 36%|███▌      | 18/50 [00:18<00:33,  1.06s/it]
 38%|███▊      | 19/50 [00:19<00:32,  1.06s/it]
 40%|████      | 20/50 [00:20<00:31,  1.06s/it]
 42%|████▏     | 21/50 [00:21<00:30,  1.06s/it]
 44%|████▍     | 22/50 [00:22<00:29,  1.06s/it]
 46%|████▌     | 23/50 [00:23<00:28,  1.06s/it]
 48%|████▊     | 24/50 [00:24<00:27,  1.06s/it]
 50%|█████     | 25/50 [00:25<00:26,  1.05s/it]
 52%|█████▏    | 26/50 [00:27<00:25,  1.05s/it]
 54%|█████▍    | 27/50 [00:28<00:24,  1.05s/it]
 56%|█████▌    | 28/50 [00:29<00:23,  1.05s/it]
 58%|█████▊    | 29/50 [00:30<00:22,  1.05s/it]
 60%|██████    | 30/50 [00:31<00:21,  1.05s/it]
 62%|██████▏   | 31/50 [00:32<00:20,  1.05s/it]
 64%|██████▍   | 32/50 [00:33<00:18,  1.05s/it]
 66%|██████▌   | 33/50 [00:34<00:17,  1.05s/it]
 68%|██████▊   | 34/50 [00:35<00:16,  1.05s/it]
 70%|███████   | 35/50 [00:36<00:15,  1.05s/it]
 72%|███████▏  | 36/50 [00:37<00:14,  1.05s/it]
 74%|███████▍  | 37/50 [00:38<00:13,  1.05s/it]
 76%|███████▌  | 38/50 [00:39<00:12,  1.06s/it]
 78%|███████▊  | 39/50 [00:40<00:11,  1.06s/it]
 80%|████████  | 40/50 [00:41<00:10,  1.06s/it]
 82%|████████▏ | 41/50 [00:42<00:09,  1.06s/it]
 84%|████████▍ | 42/50 [00:43<00:08,  1.06s/it]
 86%|████████▌ | 43/50 [00:44<00:07,  1.06s/it]
 88%|████████▊ | 44/50 [00:46<00:06,  1.06s/it]
 90%|█████████ | 45/50 [00:47<00:05,  1.06s/it]
 92%|█████████▏| 46/50 [00:48<00:04,  1.06s/it]
 94%|█████████▍| 47/50 [00:49<00:03,  1.06s/it]
 96%|█████████▌| 48/50 [00:50<00:02,  1.06s/it]
 98%|█████████▊| 49/50 [00:51<00:01,  1.06s/it]
100%|██████████| 50/50 [00:52<00:00,  1.06s/it]
100%|██████████| 50/50 [00:52<00:00,  1.05s/it]
Finish generation in 54.51199126243591 secs.

This output was created using a different version of the model, jyoung105/auraflow-v3:91da0bd7.

Examples

View more examples

Run time and cost

This model costs approximately $0.032 to run on Replicate, or 31 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 33 seconds.

Readme

This model doesn't have a readme.